I'm not sure that there is a straightforward way to do this in QIIME2. Filtering your feature table by metadata groups and then summarizing may just be the best way to do this.
@colinbrislawn's suggestion to export and open, e.g., in excel would be the only way to really pick it apart sample-by-sample if that's what you want to do (unless if you are comfortable with python programming and could access/summarize this file via QIIME2's artifact API). You could convert to a presence-absence feature table, which would make counting samples in excel easier (then you just sum the row to get the number of samples).
Also check out this thread. It sounds like your idea is a bit different, but maybe not — having multiple users with the same need would help prioritize a new feature.